Patent analysis tools though all geared to deliver information and intelligence around patents are often different in terms of their features and capabilities. How do you know what features to look for in a solution that would help your organization? Here is a brief overview of the various important features found in most comprehensive patent data analysis tools and  what components to look out for during the evaluation process.

Search Capabilities

In most cases the search has already been done on a database, so why would you want search in an analysis solution? Searching is usually a requirement for custom portfolio categorization and further its critical when conducting deeper analysis such as infringement or FTO. Even when looking at a chart, its important to be able to immediately drill-down from the chart and search through the segmented portion to get to the answer or insight you seek. Finally, many a time you have your custom fields such as  Docket Ids, tags, comments etc associated with each record and you would like to include these in your search.

Patent Text Mining functions

When working with sets of 10000 records or more if you want to quickly understand what locate topics of interest or sometimes even understand which topics are prominent across the set and what sub-topics they are associated with, a powerful tunable auto-categorization engine is a must.Solutions that are created specifically for mining and clustering patent text are better suited than more general text mining solutions. Generating keyword lists and using such list for statistical analysis is another popular method used to gain insights on trends across companies, inventors and their patents.

Analytics Capabilities

Some of the capabilities that should be covered are co-occurrence matrices, generating top 10/20/50 lists and  citation analytics. Together they help analyze relationships and spot trends within a certain space and lifecycles of technologies and more. Drilling-down to the actual patents from any of the functions is key.  It’s always good to get a few analytics questions that your organization has, tested and verified using the analytics capabilities so that you are sure if the solution is up to the task of handling your patent intelligence needs.

Charting and Visualization

Converting the results from patent analysis into a powerful visual or graphical representation with ease is a definite must. In many cases exporting visuals to image files is  critical so that they can be reused in your reports or power-points. A good visualization and reporting component helps convey the insight easily without any need for explanation. Powerful visuals necessitate actions from the information consumers typically senior management and result in swift decision making.

Reporting

Generating different styles of reports to suit varying needs across an organization is perhaps the most common activity. Support for a wide range of flexible report types is useful with output format as word or excel so that the user can modify the report as per needs after it has been generated.

User Interface & Ease of Use

There is always a fine balance between having a feature loaded analytics software tools  packed to the brim with options and one that is easy to use even for  someone with relatively little experience of patent information. A good solution in this respect is one that has a relatively easy to use user interface which one can get familiar with in a short time and at the same time have the flexibility and options to allow the user to customize features, options and the environment that they will be working with regularly when any patent analysis has to be done. A very easy to use interface may not always give you that flexibility and range of customization while an extremely complex interface may defeat the purpose and make the process of analyzing data more tedious rather than easier. Look out for an interface which gives you both ease of use along with flexibility of options.

Apart from the above some of the other capabilities to look for are data export features, associating your own custom fields with patent data, rating or scoring system and performance at 10000+ record levels.

        

A common challenge we have seen across organizations who have access to reliable IP information and intelligence from various patent databases but find it overwhelming to integrate, manage large volumes of patent data, organise various ongoing research and investigation projects and to quickly locate answers to their questions. The solution to this lies in having a comprehensive patent research and analysis platform which can make it most of your activities much simpler. Here is a 60 second Slideshare presentation which highlights this challenge. Be sure to catch the link to our 7 minute webcast video at the end of this presentation!

 

        

It’s everywhere! Blogs are buzzing with opinions and views on it, Twitter is flowing with tweets and references to it and IP professionals, innovators and the software industry are glued to  In re Bernard L. Bilski and Rand A. Warsaw v. Kappos  where the battle for patent rights for software and in particular business methods has turned into a war of sorts.

Some of the large stalwarts in software such as IBM, Yahoo, Accenture among many have shown strong support for extensive patent rights for software.  Several other smaller software firms, developers and organizations like the Free Software Foundation see software patenting as detrimental to progress and future of the software technology industry. Gene Quinn of IPWatchDog.com in his post went a somewhat fresh route and not just backed the need for software patents but also insisted that star-ups and smaller companies have a lot to gain from filing patents for their software technology in his recent post  “Why All Small Businesses Need Software Patents”. The battle on whether software patents should be allowed or not is a tough one with arguments both for and against this idea.

Those against it argue software code is logic and mathematical and the current patent system is just not equipped to handle such an issue. It’s almost like Twitter can make a claim they invented the “What are you doing now?” component of the application which allows people to update what they are doing through their software and claim “this is code” they used and it should be protected from others using their idea. Someone like Facebook however could claim that they already had this feature as part of their application where people could update what they were doing and share it with friends long before Twitter and their code though different was the original claim to this innovation. The arguments can go on endlessly and further innovation and development of software can be hampered if every single software component some developer programmed were to be patented and off-limits to other software not to mention how much more expensive software could become if every such component had to be licensed to create a working application.

On the other hand there is very little advantage for a true innovator or first mover in the software field and very little protection from the large players. For example if a smart developer invents an algorithm which can help identify and recognize objects and details in graphic images like photos and wants to monetize this through building a full application and creating a start-up, it could be just a matter of time before some large player like Flickr or Microsoft discover this and have their R&D departments replicate the technology perhaps with different code. In this system, those with the marketing might or large pockets to buy smaller players benefit without the real innovators being rewarded proportionately. That alone makes a fairly strong case for software patenting. 

Whichever side you take, there are compelling arguments and there is no doubt that if a race for patenting software inventions starts, it will have a long term impact on the industry. It could change the way we develop software or at least the approach development. It would mean rather than simply checking for copyrights and licenses for certain applications or components, businesses will now have to study and analyze all patent sets around their project to look out for liabilities and possible infringement cases before getting behind the screen to start coding. It’s still important to consider the software industry can’t easily be compared to the pharmaceutical industry or the automobile industry where patenting and IP protection have been an inherent part of their world. Yet, this new industry has developed at lightning speed and the stakes are high for everyone involved. Though this battle is not over, the way it finally turns out will depend heavily on public view and everyone’s opinion.

So what is your take on it? Should extensive software patenting rights be granted? How would you see this battle resolved?

        

Management of technology and research relies on being able to make the right decisions at the right time which in turn depends on the quick availability of reliable information. Whether deciding on strategy like what direction the innovation and research arm of the business should take and which are the areas the competition has left room for improvement on to tackling issues like finding the right people to develop a certain technology or to license a component from, the management responsible for products, research, development and technology need a constant birds eye view of not just their own IP assets but also a clear perspective of the competition and the wider technology landscape to steer the business in the right direction. This landscape however is constantly evolving with others in the market also on a quest to innovate and develop and protect their intellectual property along the way creating the need to have access to fresh information promptly when needed.

From what we have observed, the common challenges that companies face with developing a robust process of IP information delivery to management for decision making are:

• They have access to a large amount of patent data in the form of database subscriptions, downloaded documents and online sources but it's unorganized and every time they need to analyze the data for a new question or requirement management asks, they need to search through the data, gather the relevant documents and pages and then locate what they are looking for which is a long process that is often repeated.
• They have a lot of patent data related to their technologies or domain but find it challenging consolidating all the right data so they can start analyzing it and present only what is relevant to the managements questions.
• They find what they are looking for after analyzing the patent sets required to answer questions and they understand the answers but find it challenging to present it quickly and simply to management in a form that they can understand as clearly. They find communicating what they see within the patent sets to others difficult.
• The time taken to process all the data and generate answers is often too long which is why they prefer going by existing reports or previous information which may be available online through searching websites rather than generating fresh reports from the latest databases on demand.

For example, let's assume a pharmaceutical company has had plans for quite some time to develop a certain drug for a specific heart disease made some breakthrough but came across delays in the process of testing and clinical trials. In the few years the drug was under testing and not ready to hit the market there is a possibility that others may have been pursuing a similar direction of research and new patents may have come into force since creating hurdles for the launch of the drug. The management can't rely solely on what they knew at the time they started the research. They need quick access to fresh information to see the road ahead at every stage of the technology management process and make decisions. That view of the technology landscape they need exists in patent databases but needs text mining and analytic support to bring out that clear perspective needed for them to make those decisions.

At any given time in the process, management may need some specific questions they need answers to quickly to make their call. What are the competing technologies in their domain? How many other businesses are interested in this technology? Which markets already have players for this technology and where are the gaps that can be filled? Where are the potential licensing opportunities for what we develop? Who are the researchers or innovators who can help us develop what we are looking to create? These are just some examples of questions management need answers for and turn to their patent information to have them answered. Through a clear cut process of:

Search & Retrieval > Clustering > Data Clean Up > Analysis > Report Generation

Vast amounts of patent data can be mined, queried and analyzed and converted into precise visual reports that are easy to comprehend in a matter of minutes. Patent analysis software which can process this allows for quick process of intelligence delivery to upper management whenever needed. As useful as the process is to management, it can prove just as valuable to scientists, innovators researchers and product development teams to present their suggestions and cases for certain decisions and being able to back it up with visual presentations and valuable insights gathered from the data.

Equipped with the right software, the challenges of being able to generate and deliver decision making intelligence to management as well as the technology and research teams can easily be overcome. We all tend to rely on what is easily available. Once the right information is made easy and available, relying on fresh reports and insights becomes a regular practice and the technology management function becomes that much more effective.

        

September 18th 2009 marked the end of a two day international symposium for IP heads by WIPO. Based on the 2009 World Intellectual Property Indicators report by WIPO which was also recently released we can only guess there were a number of important issues to discuss including the impact of the global economic crisis and recession on the filing of new patents across patent offices around the world. Though the report doesn’t  indicate exact figures, it does mention there was a decrease in the total number of patents filed for the year 2008 although over 760,000 patents were issued during the year and 1.85 million patents filed. Surprisingly, ASIA was more resilient and has been seeing steady growth in South Korea and China which grew 23.9%.

The patent statistics within the report create an insight into the global trends in terms of innovation and highlight several interesting facts around patent activity around the world. This information is particularly interesting for us at Patent iNSIGHT Pro since we work with patent data analysis, patent reports and statistics every day. These are the kind of findings which give businesses and innovators the insight needed to know what is going on around the world of intellectual property. It helps draw visible conclusions based on analyzing thousands and even millions of patents and present the findings in a simple easy to understand reports and graphs. Each finding implies something  useful for some business or innovator around the world based on their field or business.  On their own, there are plenty of interesting facts published in this year’s report which you can read here. As the name of the report suggests, the value lies in what the findings of the data analysis indicates and what it implies to your business.  Some of the interesting statistics published in the 2009 WIPO report with regards to patents include:

"The five largest patent offices (China, European Patent Office, Japan, the Republic of Korea and the United States of America) accounted for 69% of total resident filings and 81.5% of non-resident filings in 2007. The high ratio of non-resident filings compared to resident filings is partly due to the fact that all patent filings at the EPO are considered as non-resident filings.

In 2007, the largest number of resident filings originated from Japan, the United States of America, China and the Republic of Korea. Between 2003 and 2007, resident filings of Japan declined by 1.8% per year. In contrast, resident filings of China and the Republic of Korea grew by 28.1% and 9.3% respectively.

There has been a steady increase in the total number of patent families during the past 15 years. The total number of patent families (based on first filing date) in 2006 amounted to 946,498, representing an 8% increase from the previous year.

During the period 2002-2005, patent applications in the fields of computer technology, telecommunications and audio-visual technology had strong growth, with the annualized growth rate surpassing 6%. In contrast, patent applications in the field of biotechnology have gradually decreased over the same period.

Medical technology accounted for the largest share of foreign-oriented patent families for Australia, Israel, the United Kingdom and the United States of America. The largest number of foreign-oriented families originating from France and Germany were in the field of transport.

In the field of telecommunications, applicants from Finland, Sweden, Republic of Korea and China have an above average concentration of foreign-oriented patent families. Applicants from Singapore and the Republic of Korea have a high concentration of foreign-oriented patent families in semiconductors. Applicants from European countries have an above-average concentration in the transport and engines, pumps and turbines fields.

For the first time, a Chinese company (Huawei Technology) topped the list of applicants with the highest number of PCT filings. Panasonic Corporation (Japan) and Philips (Netherlands) were ranked second and third. US universities dominated the list of top PCT applicants for the university sector. The University of California filed 345 PCT applications. Tokyo, Seoul National, Imperial College and Osaka are the four non-US universities in the top 20 list."

All these findings may be viewed to imply something to various businesses and innovators.  To a business located just about anywhere in the world which is looking to protect their IP interests in markets around the world the first finding on the offices with the most non-resident patent filings may imply that these five are top priorities for filing new patents going by the general trend of others. Based on this report they may want to initially file patent applications within these markets to start off with and then review the other countries which accept less than 20% of the total applications.

Similarly, a business or innovator working within the telecommunications technology space, may want to first protect their IP in countries like Finland, Sweden, Korea and China being above average concentration of foreign-oriented patent families. The same report also implies that there could be a much greater potential for licensing patents within these countries for anyone holding patents for telecommunications technology and displays revenue opportunities in these countries for their innovations.

The WIPO database holds a wealth of information which could be extremely valuable for smarter well informed decision making for businesses. This particular report shed some light on what is going in general with IP around the world. A macro view so to speak but there is so much more the data can tell with regards to highly specific areas of interest. With the right analysis software tools you can look into useful facts that pertain directly to your area of interest and gather intelligence that helps make better decisions. The database of over 63 million patents which are already in force around the world is your ocean of data to explore. It’s up to each one to discover what are the indicators they would like to track based on their IP intelligence needs

        

Universities in the US encourage researchers with innovative minds to pursure their inventions and invest into patenting promising inventions expecting to generate profits by licensing the developed technology to interested companies. The technology transfer offices of the universities usually get many invention ideas from their researchers and have to decide which ones to invest further time and resources into. Many have already build valuable patent portfolios as a result of academic research and innovations which have become continuous sources of licensing revenue most of which is a lot more that research and legal investments made and so is used to further catalyze the innovation assessment and investment process. Here is an excerpt published by Forbes in a 2008 article on Stanford University:

Stanford University's fertile breeding ground for breakthrough technology may have spawned the likes of Hewlett-Packard and Google, but little Stevens Institute of Technology in Hoboken, N.J., really knows how to get serious returns on its research and development.

To wit: In 2006, the school took in $4.5 million in research-related income (including licensing revenue and returns on equity stakes in start-ups) while shelling out $28 million on research--a 16% yield. That same year, Stanford pulled in $62 million against a $700 million investment; return on investment (ROI): 8.7%.

While perhaps thousands of universities globally develop valuable IP during the course of their academic research, very few file patents to protect it and fewer still are able to effectively capitalize by licensing their patents. Many US universities hold multiple patents that have failed to find licensees. Many of these are presently marketed by publishing invention details in journals, industry and trade newsletters, conference magazines (AUTM), using agents or brokers. While the culture of investing in research and filing patents has developed and matured in many US universities, being able to spot revenue opportunities within the patent portfolios already held by universities can be accelerated with strong patent analysis support. Much of similar kind of analysis is undertaken by Patent Licensing and Enforcement Agencies (See previous blog) and IP brokers.

Many universities expect not all their patents will be licensed. In fact most rely on a few blockbuster patents to bring home the revenue. Usually only 0.6% of licenses generate in excess of $1,000,000 in annual royalties.

Clearly there is need for universities to be more aggressive on the patent portfolios held by them. Assertive licensing and identification of infringers is important. However a change in research approach is required and universities need to analyze the technology areas their inventions lie in and gather data that can be used to back  their assertive licensing strategies. By improving their internal  patent analysis activities, universities can:

• Understand the market place and discover licensing opportunities and spaces in the market for research and technology even before filing for patents
•  Discovering licensing revenue opportunities and spotting companies that are infringing their IP or those who would like to

In short, these universities can use IP intelligence to work smarter , make fewer more lucrative investments in research, targeting select research areas, knowing where the opportunities lie and filing for patents specifically for IP which they are more certain will yield revenues. Good analysis will effectively help achieve better ROI on their IP investments. Even if a small percentage of the Universities out there follow in the footsteps of the those with large patent licensing revenues, the impact will be big. Only time will tell if they do.

        

Most companies hold on to their patents as a defensive means to protect themselves from competition gaining an upper hand from their innovations and development efforts. Their ability to drive maximum revenues for the business comes from their ability to have exclusive access to their innovation and fend off the competition for a stipulated time giving them a clear head start. Then, we have the Patent Licensing & Enforcement Companies (PLECs) who either acquire patent portfolio or partner with the patents holders and then look to license their technology with a completely offensive strategy. They drive revenues from assertive licensing via legal notices or even going to the extents of filing law suits against those who infringe on the patents. Due to the nature and amount of damages being awarded by the courts it may appear that Patent infringement cases can be even more profitable and less cumbersome than manufacturing products and getting them out to market. So it’s little surprise that this is a growing business model. Intellectual Asset Management Magazine online stated in a 2008 article that the IP awards and settlements market in the US is $3.4 billion annually.

Small innovators can see PLECs as a necessity to help them license their patents and take on a large companies who may be potential infringers while traditional R&D houses see these as Non-practicing Entities (NPEs) or in some cases as Patent Trolls i.e., those that buy up patents only to benefit from suing companyies having established products in the market. While there are both arguments for and against these intermediaries, they are very much a part of the innovation industry as licensing as well as return on patent investments become a concern for many. As long as businesses and inventors will need help in these two areas, there will be a market for those who can help provide this support.

Whether you are on the side that is looking to license and protect your patents or the side that needs to steer clear of PLECs, patent trolls and other businesses who may have patent portfolios that overlap yours, your intelligence will primarily come from patent databases in addition to secondary online information such as company and products literature.

In a previous post titled Unlocking value from your IP using Patent Citation Intelligence  we discussed at length ways to discovering licensing options to maximize revenue. Searching for infringing companies is not too different from searching for potential licensees. The process of assertive licensing usually involves demonstration of infringement and so technologies present in your patent sets need to be matched with either the claims of the other party or the infringement must be evident from the products functioning or product literature. Tools to fast search, lookup, rank claims help save a lot of time as part of this process.

Analyzing patents for infringement and being able to navigate around potential infringement suits are two sides of the same coin where information in patent data can come to the rescue. For both parties who need to enforce patents and those that need to safeguard their development activities against potential law suits, you should create a list of those working in the same domain who may have a need for the technology in question or those who may benefit from using the technology. The same set of companies form a watch list which PLECs analyze closely to look out for a infringement.

A 2006 EETimes article speaks about how a small company avoided going bust from patent enforcement:

Patriot Scientific Corp. had spent nearly a decade trying unsuccessfully to establish a new microprocessor architecture when it decided it needed to do some soul-searching. It hit paydirt when that process revealed its real products: patents.

The six-person company netted more than $24 million in 2005 from Advanced Micro Devices, Casio, Fujitsu, Intel and Hewlett-Packard by licensing seven U.S. patents it considers fundamental to CPUs. And it's just getting started.

"Hundreds of companies have been put on notice as potential infringers," said David Pohl, CEO of the Carlsbad, Calif., company, which hopes to collect royalties on sales of all microprocessor-based systems--sales that are estimated at $200 billion a year. "Virtually every electronic product that a consumer or business comes into contact with is touched by this portfolio."

The extract above demonstrates how important it is to be able to identify potential licensors for a patent as well as knowing who potential infringers are- an activity that has been mastered by PLECs. While identifying companies such as AMD, Casio, Fujitsu, Intel and HP may be more obvious, being able to identify and list the hundreds of smaller companies which in this case could benefit directly from the microprocessor architecture requires an even more efficient process. Whether relying on a third party company to which the task of enforcement is delegated or done in-house, the ability to create a solid watch list and keep a close eye on infringement activity relies heavily on having good intelligence at the push of a button. Being able to view graphical representations of relationships by citation data, being able to track the use of a certain technology within various industries and also to compare phrases within the patent text of a group of similar patents to detect similarities help facilitate the process.

As the ecosystem of patents filed grows and the list of competitors or possible infingement liabilities is on the rise , it can become increasing difficult to track all the developments and cover all bases without the support of good analysis tools. Luckily patent data analysis tools have advanced in leaps and bounds giving businesses the tools needed to protect their business interests through protecting their most valuable assets - their IP. Licensing and enforcement activity is todays vanguard for a IP portfolios held by businesses and we can expect to continue seeing developments in this space.

        

Historically, the interest in unlocking patent value by licensing has picked up momentum when IBM declared more than 1 Billion USD revenues purely from patent licensing.  This is why even after having successful products, companies with 50 or more active patents are looking at ways to unlock greater value from their IP portfolios. Nowadays, dedicated licensing teams are common across all technology driven mid to large sized organizations.  

Seen in black and white from a business’s perspective, patents are filed to protect intellectual property and buy limited time to capitalize on the innovation. It offers some protection from competition and provides some space for the business to get it’s products out to market as soon as possible and make the most of what is developed. Within the stipulated time frame if a business isn’t able to capitalize on the patents, they could virtually lose out on their advantage which is why it makes sense to license the patents while they are active.

The raw material for effective licensing comes from patents itself. The balance information is mostly available on the internet. Patent Intelligence plays an important role when it comes to being able to quickly locate possible out-licensing opportunities.

So what are the different ways to go about gathering intelligence ?

The licensing-101 method is to look up the forward citations and then further analyze who owns those patents and then screen the patent claims to see if the innovations are incremental over your patents.

A more powerful and effective method is to look at more than just the forward citations and build a more comprehensive licensing set for analysis. This involves going multiple generations forward (atleast 2-3 generations) and going at least 1 or 2 generations backward and then going forward from them as well. Invention companies and universities, that invest in actively locating licensing candidates for technology IP held know the value of why it important to go back and then go forward. Usually when you do that you are in a better position to locate alternative areas of science where your technology may be applicable. This is also necessary if you have a technology  in search for a problem.

So as part of this method, if you have a source portfolio say P and you want to undertake a more effective licensing analysis then you should build a portfolio that contains (g+1)((g-1)P, (g+1)((g+1)((g-1)((g-1)P)))) where (g-1)P is a set of all backward citations of P and (g+1)((g-1)P) is a set of all forward citations of that set. Add to this atleast 2 forward generations of P and  atleast one generation back from the immediate forward generation i.e., (g-1)((g+1)P) and you now have a more effective set for analysis.

Typically it is not uncommon to run into a licensing set that has between 4000-10000 records. And once you have such a set, you can proceed with the next phase of analysis required to locate most relevant licensing candidates. This would include:

• Advanced keyword searching though the licensing set with highlighting across full/text and claims
• Similarity searching (quickly comparing overlaps in text portions between records)
• Clustering is important since many a time you are not sure about all the technologies or keywords you want to  search. Its better if important topics both big and small were directly located across the set and presented to you to review and relate. This is exactly what clustering does.
• Co-citation based clustering (i.e., grouping together documents that’s share a high level of common forward and backward citations) of records to see which are the most technologically linked records with your portfolio
• Ranking and marking patents as you proceed with identification (to avoid going through the same record twice)
• Classifying the licensing set across various US/IPC and ECLA classifications and even your custom categories to see the spread of the records in different product or business lines
• Finally you will also have to update the latest ownership information of the patent. US Assignments information can help to an extent, but in  many cases companies try to hide the ownership by having obscure holding company names and private LLC’s that are setup as a vehicle to own and operate the patents. A fair guess would be that 30% of all patents fall in such type of ownership. Ironically these are usually the more relevant lot. So in such a situation, once you discover that they are good licensing potentials then further time needs to be invested in online research, inventor look-up, corporate tree and any other method to pinpoint the exact ownership.

As you can see that because of the sheer number of records its useful to have the licensing set in a medium that helps you search or slice-n-dice the information quickly. The time factor is always important and being able to get quick patent intelligence can play an important role in identifying opportunities before others, being more informed and thereby being able to negotiate better too. Patent data analysis software can help immensely in your ability to efficiently go through large volumes of patent data and quickly get to the answers you need. You can also go a step further by analyzing patent families across countries, to understand  geographical spread and IP investments of potential licensees. This way you can rank potential partners based on the markets they have a strong presence in.

Software can help analyze complex citation relationships traversing a sequence of generations to help understand a company’s resources and value as a potential partner or licensor. The cost of filing and managing patent IP portfolios is a sizable one and ensuring maximum returns on this is the goal of every business. Investment into analysis tools like Patent iNSIGHT Pro can deliver to a business can help uncover hidden revenue opportunities and discover new revenue streams for underutilized IP portfolios and in the process pay for itself several times over. Good intelligence leads to better opportunities or at the very least, helps one see them.   

        

Have you ever tried an internet search on Google or Yahoo hoping that there will be some results on what you were looking for only to find there were so many results that you found yourself overwhelmed? Overload of information is a common and consistent challenge. The same applies to innovators, researchers and those who work with patent databases and need to analyze information. On the plus side, there is virtually unlimited access to a vast amount of patent and IP data in the form of online databases and other technical publications. This abundance of data is also the downside for many as it can prove very challenging to find specific information across a very large group of patents.

Most sources of IP data and patent information provide search capabilities and while this is an important component of any database, it’s not necessarily the most efficient way of finding what you are looking for in patent data. It’s important to understand the difference and benefits of both “search” and “text clustering”. Searching is necessary, however since searches match keywords and identify results based on the hits they can end up returning too many records many of which may not fall in the context of what you’re looking for.

Text clustering technology however, identifies meaningful clusters of text or segments of information within the patent data which is more along the lines of how researchers would look through the data. It scans, identifies and then ranks relevant topics or concepts within the data which can help the researcher interpret the information better. By looking through a generated set of topics from the search results the user can quickly identify those he would like to set aside for deeper review and those he would like to ignore or mark irrelevant. That’s because clusters can represent both: topics you want and topics you don’t want. In either case you are rapidly narrowing down your search to the relevant few.

One must however set their expectations right since there isnt (and never can be) ‘the one set of right clusters’ for a set of patent records. Most solutions that provide clustering capabilities do not give any flexibility to the user to tune the way clustering is done thereby keeping the clustering process as black-box and not allowing any refinement in the generated set of clusters. That’s because it is assumed that the generated set of topics can only be used for trend analysis and not for exploring a.k.a “digging through” or narrowing down a large patent result set. However with the right set of tuning parameters a user can quickly instruct the clustering engine to focus on the “broader topics”,  or just the “finer topics”, or to keep broader topics at first level and finer topics under them, or to give more weightage to topics or concepts containing a particular set of words. With such flexibility a user can now run the clustering engine more than once, each time with a different setting, to rapidly dissect a large patent set and comprehend its various facets. This flexibility has been the cornerstone of the text clustering capabilities provided in Patent iNSIGHT Pro and a wide range of parameters can be tuned to influence the clustering process at each step.

Below is an example of the use of this technology taken from our White Paper on Text Clustering

In this sample set, we did a simple search for the word “skateboard” in Title, Abstract and Claims of patents across key countries and then de‐duplicated the results to only unique families. This resulted in 552 unique inventions.

Text clustering was then performed using Patent iNSIGHT Pro* over the Title, Abstract and Claims sections of these patents and the results obtained are illustrated below. We have used the sub‐topics on Skateboards used in Wikipedia as a sample for cross‐reference.

For more download the white paper on Text Clustering HERE

The results are automatically categorized making it easier to narrow down on a category or set of patents and the data retrieved for analysis is far more refined. In effect, the way to better efficiency in managing larger amounts of patent data and being able to analyze the information quicker lies in the automation factor of text clustering technology.

Searching through an IP database, reading through the text of the hundreds of results and then analyzing the information manually would not only be slow but very tedious in most cases. While the benefits of smarter patent data analysis software go beyond this, for helping one find the information they need and presenting patent data in a clearer light, it’s an invaluable investment with visible returns. So while you build on the large sources of IP data you have access to and gather more data, also explore the right software tools that will help you quickly narrow down and get the most from your data. With the right patent data analysis software, even a 30,000 search result set can be managed efficiently without being overwhelmed by the volume of data.

        

A friend of mine who is an avid blogger shared an interesting story of how he landed himself in a tight situation just last week.  He received an email over the weekend from the owner of a copyrighted image he used on his blog without his permission and was now threatened with being sued and facing a fine for violating the copyright laws mentioned on the owners website. This came as a big surprise since he found the image on popular photo sharing platform Flickr, contacted the owner to seek permission, confirmed the photo was published under the “creative commons license” and made sure he complied with every requirement including giving credit to the owner with a link to the original photo and yet now he finds himself in a soup. It turns out, although he sourced it from Flickr respecting every rule and following the right process, the photo was originally copyrighted on another website and another photographer had claims on the image. Despite all precautions and doing nothing wrong, there was practically no way knowing how many people had claims on that image or coming across their websites and tracing the path of the ownership sources. 

Luckily for him, it was a clear case of him being lead to believe the image was free to use just because someone else had published it under that license but our discussion on the fiasco brought about an interesting point. He said to me “I contacted the person I thought was the owner and did everything right, how was I supposed to know who had claims on the image before he published it and perhaps who owned it even before that guy?”

Now that’s where we can draw a parallel between his situation and the thousands of us who work with intellectual property and patents. Patents, unlike copyrighted images, have citations and there are several situations where innovators and researchers need to take the time to research patent citation history just to make sure they do not infringe on existing patents. Yet, many a time they find themselves in a sticky situation much like that friend of mine where a competitor or (worse) a Non-practicing entity (NPE) / troll sues them for infringement.

Today, having a strong portfolio of patents behind a successful product is good but not enough since there still lies the threat of NPE’s and trolls whom you cannot counter-sue. Companies with a portfolio of patents in a technology area must look at backward citation mining in addition to regular search when conducting infringement/FTO analysis for their own patents in order to identify risks to their portfolios. Going one generation back ‘(G-1)P’ from your portfolio ‘P’ is not enough and you must go at least 2-3 generations back or (G-1)(G-1)(G-1)P and then go forward from there. A comprehensive backward citation research would perhaps include {(G-1)P,  (G-1)((G-1)P),  (G-1)((G-1)((G-1)P)),  (G+1)((G-1)P),  (G+1)((G-1)((G-1)((G-1)P)))}. The same technique also applies to Invalidation Research and can help locate critical invalidating prior art for a blocking patent owned by a competitor or someone else.

In citation research, patent volumes tends to quickly become unmanageable if you are working with a larger group of patents and looking through all the data for specific relationships. Citation analysis is one such area where these relationships need to be seen clearly so that nothing is overlooked and the researcher has a very clear picture of everyone who has patents and claims on anything that they are working closely with. Patent data analysis software such as Patent iNSIGHT Pro comes with citation analysis components which can create citation trees and clearly display those links in a graphical format which is easy to interpret. Using both forward and reverse citation graphs one can see clear relationships across a group of patents and the chances of missing important citations is greatly reduced. Multi-generation citation sets can be created from a starting point which itself can be a single patent or a whole portfolio. These citation sets can be compared easily via intuitive tables and charts to help both quantitative analysis and a more minute claims analysis.

Apart from infringement or invalidation research, getting an overview of the history of and invention across a time line and following its evolution and usage can provide valuable R&D insights that can help make better product and research strategy decisions while also making clear the pattern of ownership and inventors associated with the work.

Despite the clear differences in copyrights and patents, citation analysis which provides insights that can help minimize unpleasant surprises like my friend had. With the right tools, it is possible to look deeper into the data and see the broader picture with a group of patents. It can help avoid any oversights and potentially expensive mistakes which were not intentional but just happened. While one perhaps can never to enough homework on IP data, it definitely pays to be as careful, calculated and informed as possible.